Mark Eifert

A discrete battery state monitoring algorithm for lead-acid batteries

An algorithm is introduced that periodically monitors the state of charge (SOC) of a lead-acid battery without measuring the battery current and identifies whether it is above or below a calibrated threshold. Because a continuous SOC is not estimated, but the output is a discrete flag, the algorithm is said to carry out “discrete monitoring”. The equivalent circuit model of the battery consists of a series of RC elements and a voltage source representing the state of charge. The problem of monitoring the state of charge of the battery can be posed as indirectly observing the state of the internal voltage source and comparing it with a threshold. The solution presented here takes advantage of the structure of the system. The equivalent circuit model of the battery is represented in state-space form, and it is shown that inferring the polarities of the states of the RC elements may be used to carry out the comparison. The polarities define discrete states, and an observer for these internal discrete states that does not need the magnitude of the battery current as an input is implemented to carry out the comparison and define the output of the monitoring algorithm. The algorithm was designed to carry out battery monitoring with minimal infrastructure, meaning without a dedicated battery current sensor. Compared to many state of the art battery monitoring sensors, it has the additional advantage of not requiring a long period of zero battery current for recalibration purposes. While the algorithm was designed for use with a lead-acid battery, it may theoretically be used for other battery types with similar equivalent-circuits representations.